As scale is one of the three lenses from which geographers observe and recognize  the world, it is of great importance to propose this concept in geographical research. All the natural or cultural phenomena cannot be scientifically observed and modelled without correct scales, thus scaling and scaling effects becoming one of the  key problems  in geography, ecology, hydrology and meteorology in recent decades. As a core issue in the scale problem, scaling of geospatial data serves as a necessary process in the scale researches. There are level theory, fractal theory, regional variable theory, Tobler’s first law of geography and so on, supporting as the basic theoretical framework in the scaling researches. Many effective scaling approaches reference to geospatial data have been developed, and different methods applied in different areas. In GIS area, vector data can be generalized by such methods as exaggeration, simplification, selection, displacement and other generalizing operations; raster data can be scaled with resampling, fourier analysis, wavelet analysis and fractal analysis, while the scaling methods of remote sensing image can be divided into three kinds, the scaling methods based on pixel, the methods based on imaging mechanism, and the image fusion methods. In this paper, we illustrate the mechanism and applications of four major geospatial scaling methods, i.e. resampling methods, geostatistical methods, fractal analysis and wavelet transformation. Scaling effects are also an important   scaling issues, with many researchers involved in the study of geospatial variations of scaling, developing evaluation criteria of scaling, etc. Though there exist some effective scaling methods, few accuracy analysis of scaling has been developed in evaluating resulting data, and researches into scaling theories, methodologies and scaling effects still need to be deepen. Therefore, it is natural trend to develop a theoretical and methodological scaling research frame, as well as to integrated multiscale spatial data with process models of different areas.